Code for paper "Patch-Level Training for Large Language Models"
active 2024-07-17 → 2026-04-12 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 635 days from 2024-07-17 to 2026-04-12. Pushes: 7 total, peak 3 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 7 total, peak 2 in a day. Comments: 7 total, peak 2 in a day. Stars: 94 total, peak 26 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| shaochenze | 11 | 7 | 0 | 4 |
| alxndrTL | 5 | 0 | 0 | 1 |
| JizhanFang | 3 | 0 | 0 | 1 |
| jyweky | 2 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue#4alxndrTL2025-03-24 09:02LR decay during the patch-level phase ?
- Issue comment#4shaochenze2025-03-15 04:59LR decay during the patch-level phase ?
- Issue#4alxndrTL2025-03-14 19:49LR decay during the patch-level phase ?
- Issue comment#3alxndrTL2025-03-11 15:27Seq len : T or T // K ?
- Issue#3alxndrTL2025-03-11 15:27Seq len : T or T // K ?
- Issue#3alxndrTL2025-03-11 15:17Seq len : T or T // K ?
- Issue comment#2jyweky2024-09-30 18:10关于第二阶段的Token级别训练细节的问题
- Issue comment#2shaochenze2024-09-30 03:46关于第二阶段的Token级别训练细节的问题
- Issue#2jyweky2024-09-30 03:37关于第二阶段的Token级别训练细节的问题
- Issue#1JizhanFang2024-07-31 11:11关于交叉熵的具体计算细节
- Issue comment#1shaochenze2024-07-31 11:03关于交叉熵的具体计算细节
- Issue comment#1JizhanFang2024-07-31 08:26关于交叉熵的具体计算细节
- Issue comment#1shaochenze2024-07-30 09:12关于交叉熵的具体计算细节
- Issue#1JizhanFang2024-07-29 04:48关于交叉熵的具体计算细节
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 94 stars here means stars gained during the window, not the repo's star count.